package CalXtalResponse

v0r21p8

CalCalibSvc

ICalCalibSvc provides a Cal specific streamlined interface to the GLAST calorimeter calibration database. CalCalibSvc is currently the only concrete implementation of this interface.

CalCalibSvc supports the following features:

• Simple interface: requires only specification of unique Cal xtal and range as input.

• Some calibration types are vectors which represent the 'knots' on a spline curve.

CalCalibSvc generates the spline objects for these types for the user. CalCalibSvc handles storage/deallocation/validation for these objects.

Once created, spline objects are retained in memory until they become invalid. At this point, the cache is flushed and new objects are created.

• CalCalibSvc can be configured to assign different flavors to each calibration type.

• Multiple instances of CalCalibSvc may be created if the user needs more than one flavor for one or more calibration types.

jobOptions

Suitable default values are currently provided for all options, all parameters are 'optional'.

Parameters:

CalibDataSvc	Name of service from which calibration data is obtained. (default="CalibDataSvc")
DefaultFlavor	Flavor for all calibration types unless otherwise specified (default="vanilla")
FlavorIntNonlin	calib flavor for integral nonlinearity constants.
FlavorAsym	override calib flavor for asymmetry data.
FlavorPed	override calib flavor for pedestal data.
FlavorMeVPerDac	override calib flavor for MeVPerDac data.
FlavorTholdCI	override calib flavor for tholdCI data.

CalResponseTools

The CalReponseTools are a collection of tools which provide a consistent interface to calorimeter response calculations.

CalResponseTools consists of 5 interfaces.

• ICalSignalTool - calculate diode signal response (in charge injection dac (CIDAC) units) for each crystal in Cal using TDS McIntegratingHitCol and IXtalSignalTool

• IXtalSignalTool - calculate signal response for single crystal / McIntegrating hit pair.

• IXtalDigiTool allows for the calculation of the digital response of one cal crystal from diode signal levels.

• ICalTrigTool generate FLE & FHE trigger response from digi or MC TDS data.

• IXtalRecTool reconstructs energy deposit intensity & centroid from digi information

Currently there is only one concrete implementation of each of the CalResponseTools: CalSignalTool, XtalSignalTool, XtalDigiTool, CalTrigTool, XtalRecTool

jobOptions

once again, suitable defaults exist for all parameters, all params are optional. see individual tools doxygen page for jobOptions

CalXtalRecAlg

CalXtalRecAlg takes the digitized calorimeter information from CalDigiCol as input, calculates the energy and position in each hit crystal and stores this data into CalXtalRecCol. CalXtalResponse package is used for the estimation of energy & position from digi info. See documentation in CalXtalResponse for details.

jobOptions

Parameters:

CalXtalRecAlg.XtalRecToolName

name of CalXtalResponse/IXtalRecTool based tool performing xtal digi->energy conversion (default is "XtalRecTool")

CalTupleAlg

CalTupleAlg generates CalTuple entries from TDS digi data. Current entries include:

• CalXtalAdcPed: pedestal subtraced adc values per channel. (BESTRANGE)
• CalXtalAdcRng: adc range selection per channel
• CalXtalFaceSignal: Signal @ each crystal face in units of MeV deposited at center of xtal

Each of these branches is a multi-dimensional array matching CAL geometry ([16][8][12][2], representing tower, layer, xtal & xtal face respectively)

• CalXtalAdcPedAllRange: pedestal subtracted adc values for all adc channels (may be zero if data is not available). shape is [16][8][12][2][4] (last index represents ADC range.
• CalXtalFaceSignalAllRange: face signal for all adc channels (may be zero if data is not available). shape is [16][8][12][2][4] (last index represents ADC range.

jobOptions

Parameters:

CalTupleAlg.tupleName	override name of CalTuple tree. (default is "CalTuple")
CalTupleAlg.tupleFilename	optional name of CalTuple file. instructs ntupleWriterSvc to create CalTuple in it's own file instead of sharing the default file w/ other modules. The default ("") will use the shared file
CalCalibSvc	CalCalibSvc specifies which ICalCalibSvc object should be used by CalTupleAlg. default is "CalCalibSvc"

CalFailureModeSvc

CalFailureModeSvc creates a list of large-scale failures in the CAL, and utilities to search the lists to allow digi and recon algorithms to ignore hits based on those lists.

It can take lists of towers, (tower,AFEE) or (tower, Controller) pairs to create the lists of dead objects. It provides a method to see if a given CalXtalId is contained in the lists.

jobOptions

Parameters:

CalFailureModeSvc.towerList	Provide a list of towers that will be made dead. Format: "tower"
CalFailureModeSvc.towerAfeeList	Provide a list of (tower, AFEE) pairs that will be made dead. Format: "tower_afee". Afee runs 0-3: x+,y+,x-,y-
CalFailureModeSvc.towerControllerList	Provide a list of (tower, controller) pairs that will be made dead. Format: "tower_controller". Controller runs 0-3: x+,y+,x-,y-.

unit_test

• CalXtalResponse unit_test thoroughly tests CalCalibSvc, XtalDigiTool XtalRecTool and CalTrigTool.
• Most tests are based on generation of Mc hits and validation of the digitization and reconstruction of that input.
• By default the test runs on a subset of xtals, but it can also run on all crystals for a much longer, but much more comprehensive test of a calibraiton set.

• Most tests are run w/ out simulated noise in digi in order to check the precision of the code. But special noise tests are run by rerunning the same input many times & checking that the distribution of the noise is correct.

• For each test, all didgital outputs are verified including LAT, FLE, FHE & ULD threholds. Also, reconstructed energy and position are checked for accuracy.

• Samples from throughout the energy range of the instrument is checked and crystals are checked at several positions along their length.

---